A Simulated Underwater Environment for Tracking Marine Mammals Using Deep Reinforcement Learning and BlueROV2
Presented at IEEE Conference on Games '23
- Linux or Windows
- Python 3.6.8
- Pipenv 2022.6.7
- Clone this repo:
git clone https://github.com/SamuelAppleby/SWiMM_DEEPeR.git
cd SWiMM_DEEPeR- Install required dependencies:
cd launchers
.\install_pip.batcd launchers
chmod +x pip_install.sh (Optional)
.\install_pip.shIf you would like to use this paper for research, please cite as:
@INPROCEEDINGS{10333168,
author={Appleby, Samuel and Crane, Kirsten and Bergami, Giacomo and McGough, A. Stephen},
booktitle={2023 IEEE Conference on Games (CoG)},
title={SWiMM DEEPeR: A Simulated Underwater Environment for Tracking Marine Mammals Using Deep Reinforcement Learning and BlueROV2},
year={2023},
volume={},
number={},
pages={1-8},
keywords={Training;Visualization;Target tracking;Oceans;Pipelines;Reinforcement learning;Games;Unity;active tracking;marine mammals;simulation environment;reinforcement learning;autoencoders},
doi={10.1109/CoG57401.2023.10333168}}
Samuel Appleby and Kirsten Crane would like to thank their supervisors: Giacomo Bergami and Steven McGough who provided invaluable guidance and feedback during the project.